(PS7-48) Role of Family Problem-Solving and Neighborhood Social Cohesion on Child Socioemotional Learning among Families with Low-Income

Introduction: Research has shown that children raised in low-income families tend to evidence difficulties in socioemotional learning (SEL) relative to their higher income peers, yet far less is known about factors that may predict variability in SEL development among low-income youth (Li et al., 2017; Comeau & Boyle, 2018). Using an approach that focuses on multiple social determinants of health, this study aims to address this gap by conducting a secondary data analysis in a low-income subsample of National Survey on Child Health (NSCH) data. This study’s focus on both the unique and combined impacts of structural factors (e.g., neighborhood resources) and neighborhood and family factors (e.g., social cohesion, family problem-solving) on child SEL aims to provide a layered exploration of child SEL development among low-income families.

Methods: Participants were 1,098 low-income families with children (MAge = 4.04 years, 51.8% male) selected from a larger national sample of 54,103 families who completed the 2022 NSCH. A confirmatory factor analysis (CFA) was used to evaluate the latent constructs of family problem-solving, neighborhood social cohesion, neighborhood resources, and SEL. Items from the topical NSCH were selected to conduct a CFA using Mplus to validate the measurement model (Muthén & Muthén, 2017). The weighted least squares mean and variance adjusted (WLSMV) estimator was used as it provides robust fit indices and standard errors of the model, which were evaluated using Hu & Bentler’s (1999) cutoff criteria. Subsequent hierarchical regression analyses were conducted to explore the associations between latent variables and covariates in this sample.

Results: The final CFA measurement model demonstrated good model fit (CFI=.97, TFI=.97, RMSEA=.07, SRMR=.06) and all factor loadings were significant. A series of hierarchical regressions were conducted, with the final model regressing child SEL on child sex, race, ethnicity, neighborhood resources, social cohesion, and family problem solving. Results of the omnibus test indicated that model fit was significant, F(8, 1089) = 51.53, p< .0001. The adjusted multiple R-squared value of 0.27 suggests that the optimal combination of these predictors explains approximately 27% of the variance in SEL. Results of these analyses also indicated that one of the covariates, biological sex, was significantly negatively associated with SEL above and beyond the effects of other predictors [b= -0.12, SE = 0.03, t(1089) = -4.6, p< .0001, 95% CI=(-0.2 – -0.08), sr²=0.014]. Additionally, neighborhood resources [b= -0.07, SE = 0.03, t(1089) = -2.6, p< .01, 95% CI=(-0.14 – -0.02), sr²=0.005], family problem-solving [b= 0.32, SE = 0.02, t(1089) = 11.1, p< .0001, 95% CI=(0.21 – 0.30), sr²=0.083], and social cohesion [b= 0.27, SE = 0.02, t(1089) = 9.4, p< .0001, 95% CI=(0.17 – 0.26), sr²=0.059], were each significantly associated with SEL.

Discussion: Results of this study suggest that various factors at both the neighborhood and family level account for significant variability in SEL among youth in families with low-income and may inform our understanding of the development of SEL in children from low-income families and subsequent refinements in related policy.